29 CFR Part 1910
[Docket No. S-012A]
RIN 1218-AA53
Control of Hazardous Energy Sources (Lockout/Tagout)

AGENCY: Occupational Safety and Health Administration, (OSHA).

ACTION: Final rule.

SUMMARY: The Occupational Safety and Health Administration (OSHA) is issuing a standard detailing safety requirement for the control of hazardous energy as a new 1910.147. This standard addresses practices and procedures that are necessary to disable machinery or equipment and to prevent the release of potentially hazardous energy while maintenance and servicing activities are being performed. The standard requires that lockout be utilized for equipment which is designed with a lockout capability except when the employer can demonstrate that utilization of tagout provides full employee protection. For equipment which was not designed to be locked out the employer may use tagout. In addition, the standard also supplements and supports the existing lockout related provisions contained elsewhere in the general industry standards by providing that comprehensive and uniform procedures be used for complying with those provisions. This standard applies to general industry employment under 29 CFR part 1910, but does not cover maritime, agriculture, or construction employment. The standard also does not cover oil and gas well drilling; the generation, transmission and distribution of electric power by utilities; and electrical work on electric conductors and equipment. These will be the subjects of separate rulemaking efforts.

The standard contains definitive criteria for establishing an effective program for locking out or tagging out energy isolating devices and requires training for authorized and affected employees. The standard requires the employer to implement the specified procedures, and to utilize effective control measures based on the workplace hazards that are encountered. OSHA expects that this standard will prevent approximately 122 fatalities, 28,400 lost workday injuries and 31,900 non-lost workday injuries a year.

This rule, 1910.147, is being placed in Subpart J of part 1910. The present 1910.147 is redesignated as 1910.150 to allow for the new section.

DATES: This final standard shall become effective October 31, 1989, except for paragraphs (c)(4), (c)(7), and (f)(2), of 1910.147 which contain information requirements currently under review at OMB. A document announcing the effective date of the recordkeeping portions will be published at a later date in the Federal Register.

ADDRESS: In accordance with 28 U.S.C. 2112(a), the Agency designates for receipt of petitions for review of the standard, the Associate Solicitor for Occupational Safety and Health, Office of the Solicitor, Room S-4004, U.S. Department of Labor, 200 Constitution Avenue NW., Washington, DC 20210.

FOR FURTHER INFORMATION CONTACT: Mr. James F. Foster, Occupational Safety and Health Administration, Room N3649, U.S. Department of Labor, Washington, DC 20210, (202) 523-8148.

SUPPLEMENTARY INFORMATION: For additional copies of this standard contact U.S. Department of Labor Occupational Safety and Health Administration, Office of Publications, Room N3101, Washington, DC 20210, (202) 523-9667.

I. Background

OSHA's General Industry standards, 29 CFR part 1910, were originally published in the Federal Register (38 FR 10466, May 29, 1971) pursuant to Section 6(a) of the Occupational Safety and Health Act of 1970 (the OSH Act) and became effective on August 27, 1971. Before their adoption as OSHA standards, these occupational safety and health standards were either national consensus standards or established Federal standards. Virtually all of the current lockout provisions in part 1910 which are affected by this standard were adopted under the section 8(a) procedure.

At the time of adoption of the original OSHA standards, there was no general, all-encompassing consensus standard or Federal standard for locking out, tagging out, or disabling of machines or equipment to protect employees when maintenance or servicing activities were being performed -- a gap that this rulemaking addresses. However, OSHA did adopt various lockout-related provisions of consensus standards which had been developed for specific types of equipment. These provisions are not deleted by this rulemaking. Current lockout-related provisions in the General Industry Standards (29 CFR part 1910) are found in the following sections:

1910.178 Powered Industrial Trucks

1910.179 Overhead and Gantry Cranes

1910.181 Derricks

1910.213 Woodworking Machinery

1910.217 Mechanical Power Presses

1910.218 Forging Machines

1910.522 Welding, Cutting and Brazing

1910.261 Pulp, Paper and Paperboard Mills

1910.262 Textiles

1910.263 Bakery Equipment

1910.265 Sawmills

1910.272 Grain Handling

1910.331 Electrical Safety-Related Work Practices

Note: See Ex. 13 for a detailed list of lockout provisions in the above standards. For further information involving the use of these provisions, refer to the discussion found in Section VI, Summary and Explanation of the Standard, addressing paragraph (a)(3)(ii).

The present OSHA regulations for locking out or tagging out machines and equipment, where they do exist, are not uniform in their coverage. Inconsistencies in these regulations exist between different equipment and industries, and between different types of equipment in the same industry. Some provisions in the OSHA standards require equipment to have the capability of being "locked out," without requiring such control to be utilized. OSHA feels that the lack of a general standard, and the incompleteness of the existing provisions, have contributed to the alarming number of injuries and fatalities that have occurred.

Since the inception of its enforcement program, OSHA, for the most part, has had to rely upon the use of the "General Duty Clause" (section 5(a)(1) of the Act) citation to ensure that employers provide safeguarding for their employees from the hazards involving the release of hazardous energy. This approach has met with only limited success, limited primarily upon the need for OSHA to prove, in the event of the contest of a section 5(a)(1) citation, that the hazard was a "recognized" hazard and that the hazard was causing or could cause death or serious physical harm. Because of these difficulties, and because of the need to fill a significant gap in the current coverage of part 1910, OSHA has been working since 1977 to gather sufficient information to enable the Agency to write a comprehensive standard for energy control in general industry.

In 1977, OSHA published a Notice in the Federal Register entitled "Machinery and Machine Guarding, Request for Information on Technical Issues and Notice of Public Meetings" (42 FR 1741, January 7,1977) (Docket S-212). In this Notice, OSHA addressed the issue of lockout or tagout, including the general question of whether lockout should always be required when machinery is not in its normal operating mode, or whether alternative methods for employee protection, such as tagout, should be permitted (42 FR 1807). The purpose of that Notice was to generate information for use in updating the OSHA machine guarding standards (Subpart 0). Respondents to that Notice generally recognized the hazards to employees when maintenance and repair activities are undertaken, and the need to use lockout or tagout to control these hazards. There was, however, a considerable range of opinion regarding the effectiveness of either a lock, tag or a combination of these devices when they are used as safeguards.

The United Automobile, Aerospace and Agricultural Implement Workers of America (UAW) petitioned OSHA on May 17,1979 (Docket S-012, [Ex. 2-3]) to establish an Emergency Temporary Standard [ETS] for locking out machinery and equipment. The petition stated that there existed a need to recognize the complexities of modern industrial equipment which use sources of energy other than electricity. It contained a discussion of the increasing need for locking out equipment from cycling without warning while it was being worked on, and related the importance of applying lockout procedures to systems using hydraulic or pneumatic power, to energy stored in springs and electrical capacitors, and to potential energy from suspended parts. Abstracts of case studies for fatalities involving 22 UAW members which were attributed to lockout-related causes since 1974 were submitted with the petitions. OSHA also received other petitions and letters in support of the UAW petition from other labor organizations, including the AFL-CIO, Allied Industrial Workers, and the United Steelworkers of America.

OSHA responded to the UAW petition on September 11, 1979 [Ex. 2], declining to issue an ETS, but advising that OSHA was proceeding to draft an Advance Notice of Proposed Rulemaking [ANPR]; addressing the subject in which the public would be invited to comment on the major issues involved in the development of a standard.

OSHA published the ANPR for a standard on lockout/tagout in the Federal Register on June 17, 1980 (45 FR 41012) (Docket S-012). In that Notice, OSHA raised issues about whether or not a generic standard should be proposed; if so; what should be the scope and application of this lockout/ tagout standard; what constituted the necessary and sufficient energy isolation methods and means, and whether there was a need for written procedures and documented employee training. There was not overwhelming support in the comments submitted to OSHA for a generic standard to cover all facets of the lockout/tagout problem. The comments did indicate, however, that a performance-oriented standard offering enough flexibility to take current work practices into consideration was desirable, and that requirements for documented procedures and employee training would have many advantages. The comments pertaining to securing energy isolating devices (the use of locks or tags) did not generate an overwhelming response strongly favoring either method. The comments received in response to that Notice were utilized in the development of the proposed standard published in the Federal Register on April 29, 1988 (53 FR 15496).

There were several other inputs into the development of the Proposed Rule: First, the National Institute for Occupational Safety and Health (NIOSH) provided considerable data to OSHA on this subject. NIOSH published a notice in the Federal Register entitled "Lockout and Interlock Systems and Devices: Request for Information" (45 FR 7006, January 31, 1980) (Docket S0-12), [Ex. 2-1]) and provided OSHA with the responses to that Notice. As part of that project, NIOSH also published its "Guidelines for Controlling Hazardous Energy During Maintenance and Servicing" [Ex. 3-4]. Other important sources of information were a Bureau of Labor Statistics (BLS) Work Injury Report (WIR) survey entitled, "Injuries Related to Servicing Equipment" [Ex. 3- 3] and two OSHA directed studies -- "Selected Occupational Fatalities Related to Lockout/Tagout Problems as Found in Reports of OSHA Fatality/Catastrophe Investigations" [Ex. 3-5] and "Occupational Fatalities Related to Fixed Machinery as Found in Reports of OSHA Fatality/Catastrophe Investigations" [Ex. 3-6]; Two further studies conducted by OSHA involved the compilation and analysis of OSHA Form 36 Preliminary Fatality / Catastrophe Event Reports [Ex. 3-7] and a compilation of OSHA section (5)(a)(1) citations [Ex. 3-8].

Of great assistance to OSHA in this undertaking was the publication on March 8, 1982, of the American National Standards Institute (ANSI) national consensus standard for lockout/tagout, ANSI Z244.1-1982, "American National Standard for Personnel Protection -- Lockout/Tagout of Energy Sources -- Minimum Safety Requirements" [Ex. 3-9]. This standard lists the uniform performance requirements for developing and utilizing a lockout or tagout procedure for the protection of employees in the unexpected energization, start-up of machines or equipment or release of stored energy during repair, maintenance, and associated activities. The consensus standard was utilized by OSHA as the primary basis for development of its proposed standard.

In July 1983, OSHA developed a proposal draft of a standard for lockout/tagout [Ex. 3-10]. This draft was distributed by associations, companies, unions and individuals which OSHA was able to identify as having an interest in the regulation. There were about 80 comments received in response to this preproposal draft. The commenters were generally in support of the effort to develop a safety standard for lockout or tagout; however, some commenters objected to the inclusion of a requirement for locking out during activities classified as "normal production operations." Comments from some sources favored the use of locks rather than tags to secure energy isolating devices while others welcomed the more flexible approach of permitting the use of locks or tags. There was also considerable comment regarding the use of an Appendix. Many commenters wanted the information supplied in the Appendix moved into the body of the standard for enforceability. Others, however, wanted the Appendix material completely removed on the grounds that reference to it by the courts in contested cases would essentially make it mandatory.

The proposed standard was published in the Federal Register on April 29,1988 (53 FR 15495). Interested persons were afforded 60 days to submit comments and/or request a hearing.

On August 9, 1988, OSHA published a Notice in the Federal Register (53 FR 29920) announcing the scheduling of a public hearing and an extension of the period for the submission of comments. The hearing was scheduled for September 22 and 23 in Washington, DC, and September 27 and 28 in Houston, Texas. The comment period was extended until September 22. On August 30, 1988, OSHA published another Notice in the Federal Register (53 FR 33149) changing the dates for the Houston, Texas segment of the hearing from September 27 and 28 to October 12 and 13.

There were 16 parties who participated in the public hearing which was presided over by Administrative Law Judge Jeffrey Tureck. During the later stages of the hearing, at the suggestion of several of the hearing participants Judge Tureck established a post hearing comment period, allowing the submission of additional data and evidence through November 28,1988, and the submission of final arguments and briefs through December 23,1988. Based upon subsequent request of several of the hearing participants, the Administrative Law Judge extended the comment period until February 6,1989. Judge Tureck certified the record of the hearing, including materials received in the post-hearing comment period on May 3, 1989.

The comments concerning the preproposal draft (Docket S-012), the special studies and other information used in the development of the proposal for this standard, the comments received in response to the publication of the proposed standard, the evidence adduced at the public hearing and the materials submitted in the post-hearing comment period, were all utilized in the development of this Final Rule.

II. Hazards

Whenever machines or equipment are utilized in industry, there are hazards not only to the employees who work with the machines or equipment but also to other employees who work or otherwise are in the immediate area. Moreover, when it is necessary to perform maintenance or servicing on machines or equipment, such activities generate additional unique hazards due to the continued presence of the energy used by the machine or equipment to perform its production function. This energy can emanate directly from a power source or can be stored in the equipment itself.

OSHA believes that failure to control energy adequately accounts for nearly 10 percent of the serious accidents in many industries. The following accidents, taken from the NIOSH report entitled "Guidelines for Controlling Hazardous Energy During Maintenance and Servicing" (Ex.4), are typical of these hazards and demonstrate the applicability of the pertinent provisions in the final standard.

1. An employee was cleaning the unguarded side of an operating granite saw. The employee was caught in the moving parts of the saw and pulled into a nip point between the saw blade and the idler wheel, resulting in fatal injuries. (Failure to shutdown or turn off the equipment to perform maintenance--1910.147(d)(2) .)

2. An employee was removing paper from a waste hogger. The hogger had been shut down, but the conveyor feeding the hogger had not been. The employee climbed onto the machine, fell onto the conveyor, was pulled into the hogger opening, and was fatally crushed. There was no energy control procedure at this operation. (Failure to document and implement an effective energy control procedure-- 1910.147(c)(4).)

3. Two employees were repairing a press brake. The power had been shut off for 10 minutes. They positioned a metal bar in a notch on the outer flywheel casing so that the flywheel could be turned manually. The flywheel had not completely stopped. The men lost control of the bar, which flew across the workplace and struck and killed another employee who was observing the operation from a ladder. (Failure to control stored energy--1910.147(d)(6 ).)

4. An employee was partially inside an asphalt mixing machine, changing its paddles. Another employee, while dusting in the control room, accidentally hit a toggle switch which caused the door of the mixer to close, striking the first employee on the head and killing him. Electrical switches to activate the machine were not deenergized and air pressure to move the doors was not shut off. (Failure to isolate equipment from energy sources--1910.147(d)(3).)

5. An employee was setting up a vacuum forming machine for a run of violin cases. He leaned over the press and accidentally activated the starting switch. His head was crushed between an air cylinder and the frame hogger opening, and was fatally crushed. There was no energy control procedure at this operation. (Failure to document and implement an effective energy control procedure--1910.147(c)(4).)

6. A trainee employee was cleaning a flour batch mixer. The employee was reaching into the machine when another worker activated the wrong switch, thereby turning the machine on. The employee cleaning the flour batch mixer suffered fatal crushing injuries to his neck. There was an unwritten company procedure for locking out during all maintenance. The procedure was not followed. (Failure to document and implement an effective energy control procedure--1910.147(c)(4); failure to train employees adequately in lockout/tagout procedures--1910.147(c)(7).)

7. An employee was cleaning scrap from beneath a large shear when a fellow employee hit the control button activating the blade. The blade came down and decapitated the employee cleaning scrap. (Failure to isolate, lockout/tagout or otherwise disable all potential hazardous energy sources before attempting any repair, maintenance or servicing--1910.147(c)(2).)

Servicing and maintenance activities are necessary adjuncts to the industrial process. They are needed to maintain the ability of all machines, equipment or processes to perform their intended functions. Additionally, erection, installation, construction, set-up, changeover, and dismantling usually must be performed with the equipment deenergized. These types of operations can present the employee with the same types of hazards of unexpected activation, reenergization, or release of stored energy, therefore, they are addressed by this standard. Similarly, lubricating, cleaning, unjamming, and making minor adjustments and simple tool changes are activities which often take place during normal production operations, but which may expose employees to the unexpected activation of the equipment or to the unexpected release of the energy stored in the equipment. All of the above activities are considered to be "servicing and/or maintenance" for the purposes of this standard.

With regard to servicing and/or maintenance which takes place during "normal production operations," it is important to note that this standard is intended to work together with the existing machine guarding provisions of Subpart O of part 1910, primarily 1910.212 (general machine guarding) and 1910.219 (guarding of power transmission apparatus). When a machine is being used for production, 1910.212 requires that the point of operation be guarded. For example, when an employee is using a table saw to cut wooden parts, the employee would be protected by guards around the blade of the saw. If the employee needs to reach into the point of operation in order to adjust the work piece as part of the production process, 1910.212 requires that the guarding protection be maintained. As long as guarding is not removed or bypassed, the lockout/tagout standard is not intended to apply to these types of situations. By contrast, using the same table saw, it may be necessary for the employee to remove a piece of wood which has become jammed against the blade of the saw. In doing so, the employee might need to bypass or remove the guard on the saw and reach into the point of operation. Although this action takes place "during" normal production operations, it is not actually production, but is servicing of the equipment to perform its production function. When such servicing may expose the employee to the unexpected activation of the machinery or equipment, or to the release of stored energy, this Final Rule will apply. If the servicing is performed in a way which prevents such exposure, such as by the use of special tools and/or alternative procedures which keep the employee's body out of the areas of potential contact with machine components or which otherwise maintain effective guarding, this standard will not apply. Thus, lockout or tagout is not required by this standard if the employer can demonstrate that the alternative means enables the servicing employee to clean or unjam or otherwise service the machine without being exposed to unexpected energization or activation of the equipment or release of stored energy.

The above mentioned servicing and/or maintenance activities are currently being accomplished in general industry with varying degrees of safeguarding or protection for employees. This safeguarding or protection ranges from allowing the employee to conduct the servicing or maintenance activity which the machine or equipment is energized and operating (virtually no protection), to requiring that the machine or equipment simply be turned off or shut down, to providing for deenergization and lockout or tagout of the machine or equipment. OSHA believes that the least desirable situation is to allow employees to perform maintenance, repair, or service activities while the machine or equipment is energized and capable of performing its normal production function. The Agency recognizes that there are certain servicing operations which, by their very nature, must take place without deenergization, such as operational testing of machines or equipment. Locking out or tagging out cannot be performed during these operations, since both lockout and tagout require that equipment to be deenergized. Additionally, this standard does not apply when certain tasks are conducted during normal production operations such as repetitive minor adjustments or simple tool changes when these activities do not increase the risk of injury to employees. Conversely, operations such as cleaning and unjamming machines or equipment are covered by this standard when the employee is exposed to greater or different hazards than those encountered during normal production operations; it should be emphasized that this rule applies to cleaning and unjamming when an unexpected activation or release of energy could occur.

The vast majority of servicing or maintenance activities can safely be done only when the machine or equipment is not operating and is deenergized; therefore, these activities are covered by the standard.

Some servicing operations do not expose employees to hazards which would necessitate that a machine, equipment or process be deenergized and locked out or tagged out. Practices such as reaching around guards during the cleaning of rollers of printing presses or the feed points or screw conveyors while the equipment is operating, violate the safeguarding requirements set forth in 1910.212, and therefore, such activities are violations of that rule.

Performance of maintenance or servicing activities on a machine or equipment that is in operation has the potential of exposing employees not only to contact with moving machinery components at the point of operation, but also to contact with other moving components, such as power transmission apparatus, and also increases the risk of injury due to the position the employee must assume and the need to remove, bypass or disable guards and other safety devices. In many cases, these activities expose the employee to the hazard of being pulled into the operating equipment when parts of the employee's body, clothing or the material or tools used for cleaning or servicing become entrapped or entangled in the machine or equipment mechanism. The use of extension tools or devices to permit the operator to stay outside these danger areas, while of some benefit in reducing direct employee exposure to the hazards of entanglement or entrapment, can in itself, result in injuries to employees. This can occur, for example, when an employee is struck by the tools or devices that inadvertently come in contact with moving machine components, and are pulled from the employee's grasp.

However, shutting down a machine or equipment usually is not the total solution to the problem. Once the machine or equipment has been stopped, there remains the potential for employee injury from the unanticipated movement of a component of the machine or equipment, or from movement of the material being handled. This unanticipated movement can be caused either by the release of residual energy within the machine or equipment or as the result of the conversion of potential energy to kinetic energy (motion). For example, residual energy can be manifested by the presence of springs under tension or compression, or by the presence of pressure (either above or below atmospheric) in systems containing gases or liquids.

Potential energy is considered to be a function of the height of an object above some datum plane. This datum plane is usually considered to be where that object would come to rest if the restraint holding the object were released, such as where the upper die in a punch press is positioned above the lower die. If the restraining device holding the upper die in place was to be removed, the potential energy of the upper die would be converted into kinetic energy (downward motion), resulting in the upper die being propelled downward, coming to rest on the lower die. This motion can cause a crushing, cutting, lacerating, amputating or fracture injury to an employee's arm, hand or some other part of the body which occupies the space between dies.

OSHA believes that the most effective method to prevent employee injury caused by the unanticipated movement of a component of a machine or equipment, or of the material being handled, is either to dissipate or minimize any residual or potential energy in the system, or to utilize a restraining device to prevent movement. This can be accomplished by moving machine or equipment components to a point at which springs are at or near a neutral state, by moving components so that liquids or gases reach or approximate atmospheric pressure, and by blocking material or components or moving them to a point of minimum potential energy (moving components to a stable, resting position).

Further, even though the machine or equipment has been shut off, and even if residual energy has been dissipated, an accident can still occur if there is an inadvertent activation of that machine or equipment. Inadvertent activation can occur due to an error on the part of the employee who is conducting the maintenance or servicing activity, or by any other person. For example, the servicing employee can unintentionally cause the machine or equipment to start by shorting across electrical switches or by accidentally moving controllers (either electrical controls or valves) into the "on" or "operational" position.

An accident can also occur when another person who is not necessarily involved with the maintenance or servicing operation causes the activation of the machine or equipment being serviced. This can occur when a person uses the wrong controller and starts a machine or equipment that the employee did not intend to start. It can also occur when a person finds a machine or equipment not operating and starts it, without knowing someone else is performing maintenance or service on it. This latter type of accident is more apt to occur when the machine or equipment is large and /or complex, and the employee who is conducting the servicing activity is at a part of the system which is some distance from or not visible from the controls. The generally accepted best means to minimize the potential for inadvertent activation is to ensure that all power to the machine or equipment is isolated, locked or blocked and dissipated at points of control using a method that cannot readily be removed, bypassed, overridden or otherwise defeated. In the case of an electrically run machine, piece of equipment or process, this can be done by going back toward the original source of the power and shutting off a main switch or by disconnecting the electrical lines. OSHA believes that this action must be followed by the placement of some safeguard to prevent the reenergization of the circuit during the maintenance or servicing. To ensure that another employee will not attempt to restart the machine or equipment or to reenergize the circuit, there must be some assurance that all other employees know that the circuit is deenergized and must remain so. This can be accomplished by the utilization of a standardized procedure for deenergizing the system; by training employees to familiarize them with the restrictions of the procedure which apply to them; and by enforcing a prohibition on another employee removing or bypassing another's safeguard. Those employees whose job require them to operate or use a machine or equipment that must have maintenance or servicing performed on it, must be aware that the machine or equipment is going to be stopped or shut down, and locked out or tagged out, and that they should not attempt to restart or reenergize it. Additional training is also needed for those employees who must utilize the procedure.

Even if all other protective measures are taken, accidents can still occur following the completion of the maintenance, repair or servicing activity, if the machine or equipment is reenergized and started before all guards and other safety devices have been replaced or reinstalled. Additionally, all tools and other foreign objects must be removed from the location and a check completed to ensure that no employees are in a place where the re-energization and starting of the machine or equipment will endanger them.

III. Accident Data

The collection of data on accidents resulting from a failure to utilize proper lockout or tagout procedures is hampered because many accidents are not reported; are reported only locally; or are reported and categorized under other causal factor categories (such as "caught-in" or"caught-between"). Incorrect or incomplete categorization is particularly true for lockout related accidents, since many of the injuries are grouped under the more commonly used classifications such as, burns, electrocutions, lack of machine guarding or equipment failure.

OSHA also recognizes that there has been some underreporting of accident data -- either inadvertent or intentional. As a result, OSHA believes that the data available represent only a portion of the total injuries and fatalities that have occurred. However, OSHA believes that the accidents which have been recorded or reported and investigated or studied as being "lockout related" provide a graphic illustration of the extent of the problem, the causal factors, the distribution of accidents in industry, and the type and severity of injuries resulting from those accidents.

There have been several studies conducted to determine the magnitude and extent of the problem. These studies were conducted by: (a) The U.S. Department of Labor, Bureau of Labor Statistics; (b) OSHA's Office of Data Analysis (formerly Office of Statistical Studies and Analysis); (c) the National Institute for Occupational Safety and Health (NIOSH); (d) OSHA's Office of Experimental Programs; and (e) OSHA's Office of Mechanical Engineering Safety Standards. During the hearing, the UAW provided detailed data on fatalities and injuries (Tr. p. H216, H253), which they expanded upon in their post-hearing submission (Ex./ 3-49). The studies are discussed in the following paragraphs.

A. Bureau of Labor Statistics Work Injury Report Study. The first study examined by OSHA was the Work Injury Report Study entitled "Injuries Related to Servicing Equipment" [Ex. 33]. This study is a compilation of reports of accidents and follow-up survey questionnaires sent out by the Bureau of Labor Statistics (BLS). The survey, conducted from August to November 1980, covered workers who were injured while cleaning, repairing, unjamming or performing other non-operating tasks on machines, equipment and electrical or piping systems. BLS identified accidents from 25 participating states, and mailed each of the injured employees a follow-up questionnaire containing inquiries about the specific details of his/her accident. There were 1,285 questionnaires sent out and 833 (approximately 65 percent) of the employees responded. Not all questions were responded to by all participants, since many of the questions related to situations which may not have been relevant to the circumstances of each injury. In some instances, many of the respondents also gave multiple responses to a single question.

Tables I through VI present tabulations of the results of the BLS Work Injury Report Study.

TABLE I: INDUSTRY DISTRIBUTION--BY STANDARD INDUSTRIAL CLASSIFICATION (SIC) MAJOR DIVISION AND COMPANY SIZE

(¹) Due to rounding, percentages may not add to 100.

(²) The total of each table represents the number of respondents answering the pertinent question(s) of the survey.

TABLE II. -- OCCUPATIONAL DISTRIBUTION

(¹) Less than .5.

Note. -- Due to rounding, percentages may not add to 100.

TABLE III. -- ACTIVITY OF TIME OF ACCIDENT

(¹) Less than .5 percent.

TABLE IV. -- CIRCUMSTANCES OF INJURIES

(¹) Due to rounding percentages may not add to 100.

(²) Because more than one response is possible the sum of the responses and percentages may not equal the total number of persons who answered the question.

(³) The two accidents which occurred after the equipment was locked out took place because (1) the lockout had been done to the wrong power line and (2 a second power line had been spliced into the wiring beyond the lockout.

TABLE V. -- TRAINING

(¹) Because more than one response is possible, the sum of the responses and percentages may not equal the total. Percentages are calculated by dividing each number of responses by the total number of persons who answered the question.

TABLE VI. -- ESTIMATED LOST WORKDAYS

B. Analysis of 83 Fatality Investigations by OSHA's Office of Data Analysis.

The second study examined by OSHA was the compilation of data from 83 fatality investigations conducted by OSHA between 1974 and 1980. This report is entitled "Selected Occupational Fatalities Related to Lockout/Tagout Problems as Found in Reports of OSHA Fatality/Catastrophe Investigations" [Ex. 3 5]. All of these accidents were identified as having been caused by failure to properly deenergize machines, equipment or systems prior to performing maintenance, repairs or servicing.

Tables VII through IX present tabulations of the results of the OSHA analysis of 83 fatality investigations.

TABLE VII. -- CAUSAL FACTORS

NOTE. -- Due to rounding, percentages may not add to 100.

TABLE VII. -- NUMBER OF INJURY

TABLE IX. -- EMPLOYEE ACTIVITY

In analyzing the 83 fatality investigation reports and assigning causes to each accident, no attempt was made to draw conclusions or inferences beyond the information contained in the reports. For example, if the employee was killed in operating machinery, unless the report stated otherwise, the cause of the accident was considered to be failure to shut off the machine, rather than a combination of causal factors such as failure to sit off the machine, failure to lockout, failure to document adequate procedures, and failure to provide sufficient employee training. Additionally, if a machine was found to be running, it was assumed that the employee failed to shut off the machine rather than that another employee restarted the machine.

C. Analysis of 125 Fixed Machinery Fatalities by OSHA's Office of Data Analysis. A Separate study by OSHA's Office of Data Analysis is entitled "Occupational Fatalities Related to Fixed Machinery as Found in Reports of OSHA Fatality/Catastrophe Investigations" [Ex. 3-6]. This study contained an analysis of investigative reports of 125 fatalities involving fixed machinery which occurred between 1974 and 1976, and which were investigated by OSHA. The primary causal factors under which the accidents were classified were operating procedures, accidental activation, lack of machine deactivation, equipment failure, and other causes.

The following is a tabulation of the results of this study.

TABLE X. -- CAUSAL FACTORS, OSHA ANALYSIS OF 125 FATAL ACCIDENTS

D. National Institute for Occupational Safety and Health, Guidelines for Controlling Hazardous Energy During Maintenance and Servicing and Study of Hazardous Release of Energy Injuries in Ohio in 1983. The next studies considered by OSHA were done by the National Institute for Occupational Safety and Health (NIOSH) [Ex. 4 and 2-80c]. In the first, fifty-nine out of a total of 300 accident reports were analyzed to illustrate situations in which adequate control of energy might have prevented the accidents. These case files were selected because they contained sufficient detail to enable NIOSH to evaluate the accidents and determine what countermeasures might have been available to prevent the accidents.

The report indicated that these types of accidents are preventable if effective energy control techniques are available. the workers are trained to use them, and management provides the motivation to ensure their use.

The following is a tabulation of the results of the first study.

TABLE XI. -- CAUSAL FACTORS, NIOSH STUDY

The NIOSH draft report, undated, entitled: "Study of Hazardous Release of Energy Injuries in Ohio in 1983" (Ex 2-80c).

This report contains information on 339 accidents which occurred in the state of Ohio in 1983. These accidents were selected because: (1) They fell into likely categories of industry, occupation, type of accident, source of injury and diagnosis of injury; (2) the worker's compensation claim narrative suggested applicability; and (3) questionnaire responses by plant officials positively identified the injuries as resulting from an unexpected energy release during equipment repair, servicing or maintenance. The report defined an unexpected or unwanted release of energy "as when a press closes on an operator's hand or when steam escapes from a broken pressure line."

The "Ohio Study" was submitted by NIOSH in draft form. OSHA is not aware of whether the study results have since been finalized by NIOSH, or whether any further effort has been expended to follow-up on its findings. However, OSHA has evaluated the draft study and has determined that few definite conclusions can be drawn from the available data. For example, most of the injuries reported in the study (70%) occurred to production workers as a result of servicing which took place during normal production operations. Although the study indicated that firms where injuries occurred used tagout, it did not indicate whether either tagout or tagout procedures were applied in situations where production employees were performing servicing work, as well as maintenance employees. Without such information, it is not possible to determine whether the tagout procedure failed in situations where it was being applied, or whether tagout (or other type of employee protection, such as shutting down the equipment) was in use at the time of the accident. In addition, the study only considered the issue of locks versus tags, and did not evaluate the other elements of the lockout or tagout programs in place. As OSHA has emphasized the adequacy of a program for the control of hazardous energy relies on much more than whether a lockout device or a tagout device issued on the energy isolating means. Therefore, the Agency has determined that the draft Ohio study raises many more questions than it answers, and that no solid conclusions can be drawn from the data provided to date. OSHA encourages NIOSH to continue its review and analysis of this study, and looks forward to receiving a final version of the study after a full evaluation and revision has been performed.

The following is a tabulation of the usable results of this study.

TABLE XII. -- TASK BEING PERFORMED AT TIME OF ACCIDENT

TABLE XIII. -- EQUIPMENT MODE WHEN INJURY OCCURRED

(I) Ten respondents did not identify the equipment mode.

F. Analyses of Fatality/Catastrophe Reports and General Duty Clause Citations by OSHA's Offices of Experimental Programs and Mechanical Engineering Safety Standards.

There were two additional OSHA studies which were conducted jointly by the Office of Experimental Programs and the Office of Mechanical Engineering Safety Standards. These studies were compilations and analyses of OSHA Form 36 reports [Ex. 3-7] and OSHA 5(a)(1) citations [Ex. 3-8], respectively.

An OSHA Form 36 (Preliminary Fatality/Catastrophe Event Report) is prepared each time an Area Office is notified of a serious accident resulting either in a fatality or in serious injury to five or more employees that necessitates their hospitalization. This report is used to determine whether or not OSHA will conduct an investigation of the circumstances surrounding the accident. Since OSHA does not receive notification of all accidents resulting in a fatality or catastrophe, the total number of Form 36 reports received does not equal the total number of workplace fatalities and serious injuries which occurred during this study period. However, OSHA believes that the causes of, and the circumstances leading to, the accidents clearly demonstrate the nature and seriousness of lockout/ tagout-related accidents.

The OSHA Form 36 study which analyzed data reported during the period 1982-1983 [Ex. 3-7], utilized a list of 443 fatalities. From these fatalities, all of which occurred in industries subject to the present regulations, it was determined that 36 (8.1 percent) would have been prevented by the use of an effective lockout or tagout procedure.

The second study [Ex. 3-8] used information developed by OSHA's Office of Mechanical Engineering Safety Standards which identified, categorized and recorded "general duty clause" (section 5(a)(1) of the OSHA Act) citations from 1979 to 1984. A general duty clause citation is issued when, during an inspection, a "recognized hazard" is detected which is causing or is likely to cause death or serious physical harm to an employee, but which is not addressed in an OSHA standard applicable to that industry.

The citations in the latter study have been broken down between maritime, construction, and general industry. The general industry citations were further subdivided to reflect the nature of the hazard which the citation addressed, such as hazardous materials or material handling. When there was special Agency interest in an industry or hazard, the citations were further broken down by industry sector (such as oil and gas well drilling)

From 1979 through 1984, 3,638 inspections were conducted which resulted in the issuance of general duty clause citations. Of these 3,638 inspections, there were 376 inspections in which the failure to control hazardous energy was cited. Hence, in approximately 10 percent of all inspections which resulted in the issuance of at least one General Duty clause citation, herein referred to as a 5(a)(1) citation, failure to lockout or tagout was identified. [Ex. 3-8]

The following is a tabulation of the breakdown of lockout citations by industry division.

TABLE XIV. -- INDUSTRY PROFILE, OSHA 5(a)(1) LOCKOUT CITATIONS

Note.-- Due to rounding, percentages may not add to 100.

At the hearing, the International Union, United Automobile, Aerospace and Agricultural Implement Workers of America (UAW) testified that there were 74 fatalities which it referred to as "lockout fatalities," which had occurred to its members between 1973 and 1988 (Tr. H253). In response to requests at the hearing, the UAW provided additional information on these fatalities (Ex. 49E). (The number of "lockout fatalities" was revised to 72 in the post-hearing submission.) The post-hearing data reinforce OSHA's determination that fatalities from hazardous energy sources involve more than simply a failure to "lock out" machines or equipment. Of the 72 fatalities, UAW reported that there had been "inadequate training" in 49 cases (68%); "inadequate procedures" in 50 cases (69%); and "adequate, but unenforced procedures" in 19 cases (26%). Although OSHA agrees that lockout provides more security against reenergization of equipment than tagout, the Agency is convinced more than ever that there is much more to energy control than the question of lockout vs. tagout. The UAW data make a strong case for the need for OSHA to provide for proper energy control procedures and adequate training in those procedures.

In the proposal, OSHA estimated, based on BLS data, that lockout or tagout related fatalities represented 7% of the total number of occupational fatalities. In their post-hearing comment, the UAW indicated that for their workers, this figure is estimated to be 26%, and that OSHA should take this larger estimated percentage into account in its projections. The UAW also argued that its data base is larger than that used by OSHA, and that it is more reliable because of its national scope and inclusion of both large and small facilities. (Ex. 49A). OSHA appreciates the time and effort taken by the UAW in compiling such data and in submitting it to the rulemaking record. At the time of the proposal, the Agency acknowledged that its injury and fatality figures were likely to be understated for various reasons. Regardless of whose figures are used, there is little doubt that the failure to control hazardous energy sources exposes employees to a significant risk, and that this standard is necessary to reduce those risks.

IV. Basis for Agency Action

OSHA believes that there exists a sufficient body of data and information upon which a reasonable standard can be based to reduce the number of fatalities and injuries resulting from failure to utilize proper and adequate practices and procedures for the control of potentially hazardous energy. This position is based upon an analysis of the accident data available to OSHA, all of which is in the docket of this rulemaking proceeding.

Most accident reports break down the relevant information in accordance with the classifications contained in the American National Standards Institute, ANSI Z16.2 "Method of Recording Basic Facts Relating to the Nature and Occurrence of Work Injuries" [Ex. 3-11]. These classifications are: The nature of the injury, part of the body, source of the injury, accident type, hazardous condition, agent of injury and unsafe act. Many accident reports are generated primarily to document the occurrence of accidents and concentrate on the information which is necessary to process workers' compensation claims. For this reason, they tend to emphasize information about the injury rather than the events and condition which caused the accidents. Therefore, most of the pertinent information identifying the nature and extent of the problem of controlling hazardous energy was gathered by OSHA by conducting the special studies referred to above. Because of the limitation on the available data, no single study in itself can be expected to provide conclusive support for comprehensive regulation of energy standards. However, the studies and other available data, when considered as a whole, clearly indicate not only the scope and extent of the problem, but also the need for a comprehensive standard. The studies are consistent in their demonstration of the causative factors involved in lockout-related accidents. and they provide strong evidence for the potential effectiveness of OSHA's Final Rule in dealing with those factors.

OSHA believes that the hazards associated with the failure to control hazardous energy are widespread. The following table indicates the distribution, by industry, of the accidents reported in the Bureau of Labor Statistics (BLS) Work Injury Report Study (WIR) and in the OSHA 5(a)(1) study citations discussed earlier.

TABLE XV. -- INDUSTRY PROFILE, BLS WIR AND OSHA 5(a)(1) CITATIONS

Although employees in almost every industrial division are exposed to the hazards associated with the unexpected energization or start up of machines or equipment, or by the unanticipated release of stored energy, the preponderance of the accidents and injuries occur in manufacturing (Division D). It should also be noted that Services [Division I] includes many employers who perform maintenance on equipment in manufacturing and other sectors covered by Part 1910.

In addition to the accidents which could occur when maintenance or servicing is being conducted, OSHA also identified some accidents which could occur while employees are lubricating, cleaning, unjamming or adjusting machines or equipment. These activities differ from other activities which are routinely conducted during normal production operations in that they are usually done only on an as-needed basis. When these activities are being conducted during normal operations, the machine guarding required by other OSHA standards (that is, ' 1910 912 for point of operation guarding and ' 1910.219 for power transmission apparatus guarding) may afford the necessary and sufficient protection for the employees performing those activities. However, in many instances the employee must either remove guards or other safety devices or work under unusual circumstances which would subject the employee to a different or greater risk than would be encountered during normal production operations. In those instances OSHA believes that the machine or equipment must, if possible, be shut down and locked or tagged out to protect the employee from injury.

As noted earlier, OSHA has evaluated section 5(a)(1) citations that were issued for failure to control hazardous energy, and has determined that this area accounts for about 10 percent of the serious hazards not presently covered by a specific OSHA standard. The seriousness of the hazard to be addressed by this standard is highlighted by the fact that section 5(a)(1) citations are issued only for recognized hazards which cause or are likely to cause death or serious physical harm. Similarly, the OSHA Form 36, also discussed above, is initiated only when OSHA is notified of deaths or multiple hospitalizations. Further analysis of the lost workday data from the BLS WIR indicates that the severity of injuries from failure to control hazardous energy sources (an average of 24 lost workdays per lost time injury) is much higher than the national industry-wide average of 16 lost workdays [Ex. 14].

In developing this Final Rule, OSHA has estimated the total numbers of fatalities, lost-workday injuries, and minor injuries attributable to lockout-related accidents. These estimates were based on an extrapolation of the available national data sources discussed earlier [Ex. 3, 5, 6, 7]. From these data the number of preventable accidents was determined. OSHA believes that the Final Rule will prevent 85% of the total numbers of injuries or fatalities from exposure to hazardous energy in the workplace. The Agency estimates that approximately 31,900 minor (non-lost-workday) injuries; 28,400 lost-workday injuries; and 122 fatalities per year (based on 1984 accident levels) will be prevented by this standard. (see Section on Regulatory Impact Analysis below). These estimates were derived by identifying the percentage of accidents in various data sources which were determined to be lockout-related and applying those percentages to the number of accidents. It was determined that two percent of all nonfatal accidents and 7.1 percent of all fatalities occurring in general industry related to failure to adequately control hazardous energy. In addition, the data indicate that the risk of accidents and injuries is independent of the number of employees in a particular workplace. This finding is predicated upon the distribution by size of the companies which employed the injured employees surveyed in the BLS WIR. In the survey, almost as many respondents (392, or 49 percent) reported that they were employed at facilities of 100 or more employees as those who were employed at facilities of less than 100 employees (402, or 51 percent).

Based upon analysis of the aforementioned evidence, OSHA believes that the failure to control hazardous energy results in a significant risk to employees. Further, the data clearly demonstrate that the consequences of an accident involving failure to lockout or tagout are more severe in terms of lost workdays than the average industrial accident. OSHA also believes that a significant risk from hazardous energy extends across many segments of general industry. OSHA has also analyzed the studies to determine the underlying causes of the conditions which existed when lockout related accidents occurred. From this information, OSHA developed a list of measures which would have prevented most of the accidents in the studies, and used this list to develop its proposed standard. It should be noted that the studies vary widely in the quantity and quality of the information provided for the reported accidents (different methods of reporting, and incompleteness of the findings of the causes of the accidents, for example). Therefore, professional judgment was used in the interpretation of the results of the studies, in order to provide a comprehensive evaluation of the data and to correlate the information on accident causation. While the numbers and percentages from all studies do not necessarily agree, the studies all indicate the existence and seriousness of the problems, and provide valuable information as to measures that are necessary to correct the problems. Tables XVI through XX below cover what OSHA believes are the major causal factors in lockout-related accidents, and indicate the prevalence of such factors as reflected in the different accident studies.

TABLE XVI. -- SERVICING ACCIDENTS OCCURRING WHILE EQUIPMENT IS OPERATING

The reasons most often given in the BLS WIR for not turning off equipment prior to servicing were that it would take too long or slow down production; it was not required by the employer; it was not necessary; or the task could not be done with the equipment off.

As pointed out in the Hazards section of this Notice, just shutting off a machine, equipment or process may not completely control the hazardous energy. Even after a machine, equipment or process is shut down, residual energy may still be present in the form of moving components, spring or hydraulic pressure, the force of items which have become jammed in machine parts, or the energy which is stored in machine, equipment, or system components due to their position (potential energy).

TABLE XVII. ACCIDENTS DUE TO FAILURE TO ENSURE POWER OFF

The Hazards section of this Notice also discussed the fact that even though the machine, equipment or process has been shut down, and the residual energy controlled or dissipated, an employee can still be injured if the machine, equipment or process is restarted by either that employee or another employee. Injury can occur when an employee inadvertently contacts switches, valves or other controllers or when an employee activates the equipment without recognizing the reason it was shut off, inadvertently exposing other employees to a hazard.

TABLE XVIII. ACCIDENTS DUE TO INADVERTENT ACTIVATION

Clearly, it is insufficient simply to shut off machinery to conduct repair, maintenance or servicing. OSHA believes that some means must be utilized to ensure that employees are safeguarded during those operations.

After servicing, there is also the need to ensure that all guards have been replaced, that all tools and other extraneous materials have been removed from the machine, equipment or process, and that reenergizing and starting normal productions operations will not subject an employee to an increased potential for injury. This is especially true when the maintenance, repair or service is conducted at or near an employee's workstation.

OSHA believes that many of the problems of de-energization and reenergization of machines or equipment can be reduced by the employer's development and utilization of a program which incorporates a standardized procedure for servicing/maintenance operations. The procedure would outline the necessary steps to be taken to prepare for, conduct and complete servicing of equipment, and the program would provide employees with an understanding of the procedure and the reasons why it must be followed. A program can provide the details to be followed in performing servicing operations safely (the procedure), together with the training and motivation needed to assure that employees understand and implement those details.

TABLE XIX. ACCIDENTS ATTRIBUTABLE TO EMPLOYER NOT HAVING OR EMPLOYEES NOT UTILIZING A PROCEDURE

OSHA believes that employee understanding and utilization of a standardized procedure are critical to the success of a lockout or tagout program. Without these elements and commitment from management, the effectiveness of the program can be seriously compromised. Proper training in the procedure, and explanation of how it works and why, are crucial to its implementation by the employees. Even though there can be no exact quantification of the effects of training employees, the BLS WIR Study gives an indication of the effect of the lack of training in the necessary measures to be taken in deenergizing machines or equipment (see Table XX below).

TABLE XX. LOCKOUT TRAINING OF INJURED EMPLOYEES, SOURCE: BLS WIR (FROM 613 RESPONSES)

Of those injured employees who had received training, 15 stated that their training had occurred after their accident. Additionally, 60 employees stated that they had received their training more than a year prior to the accident. Even though training has been provided at some time during employment, the length of time between the receipt of the training and the accident is a limiting factor on any beneficial effect that has been derived from the training. In the Final Rule, discussed below, OSHA recognizes the need for remedial or refresher training of those employees who must use the procedure, and that such retraining must be conducted at least annually.

Based upon an analysis of the rulemaking record. OSHA believes that the safe performance of activities such as repair, maintenance and servicing. requires the deenergization of machines or equipment whenever feasible. Further, in order to ensure that maintenance or servicing activities are conducted safely, a lockout or tagout procedure must be utilized. This procedure must call out the steps to be taken to deenergize the machine, equipment or process; to ensure that the deenergization is sufficiently complete; to dissipate or prevent the release of residual energy to ensure that the machine, equipment or process cannot be reenergized accidentally or unexpectedly: and to ensure that the reenergization is accomplished safely. The establishment and utilization of the procedure must be coupled with sufficient initial and follow-up training to ensure the successful utilization of the procedure. Next>>

| Parts I-IV | Part V-VI | Parts VII-XVI |